Comments on: Narrative of education in Pakistani media sources https://partnews.mit.edu/2014/04/02/narrative-of-education-in-pakistani-media-sources/ Treating newsgathering as an engineering problem... since 2012! Wed, 02 Apr 2014 15:55:45 +0000 hourly 1 https://wordpress.org/?v=5.2 By: Erhardt https://partnews.mit.edu/2014/04/02/narrative-of-education-in-pakistani-media-sources/#comment-29713 Wed, 02 Apr 2014 15:55:45 +0000 http://partnews.brownbag.me/?p=4688#comment-29713 Great bit of media ecosystem analysis! Unsupervised learning algorithms are not a normal part of the data journalists toolkit but I expect they will be in the future. They will certainly play an ever larger role in news discovery that feeds stories for journalists to write as we see in The AOL Way approach.

I like the bubble graphs and the two-dimensionality of size and color, but the third dimension of spatial orientation is distracting. Obviously, in the first graph the spatial orientation is in part representative of clustering, but that’s already covered by color. What I don’t understand is the core, periphery, and outer periphery nature of it. Is there something about the core that is different and worth paying attention to because the layout makes me think so but you don’t mention it.

In the subset bubble graphs, I like that I can focus more on the words (they are bigger), but I find the colors distracting. What do they mean? Can you explain or offer a legend of some kind?

The final timeline is great! I love the trendline and the labeling of the spikes. You should be consistent about either pointing to the spikes or giving a specific time for each event like you do for Malala’s speech so that we know exact what it refers to.

Overall, very interesting!

]]>